A pediatric psychologist, a behavioral pediatrician, and a pediatric oncologist working as a team will study children and adolescents with cancer to determine: (1) the relative efficacy of hypnosis, supportive counseling and assessment alone on the reduction of chemotherapy-related nausea and vomiting, (2) if any beneficial effects achieved with active therapist-directed intervention are maintanined by the patient without the therapist, and (3) the relationship between hypnotic susceptibility and symptom reduction for those patients receiving hypnotic intervention. Patients' and parents' ratings of the severity of nausea and vomiting and the duration of these symptoms, including disruption of eating, sleeping, school, and activities, will be obtained immediately following each of two matched (for drug types and dosages) consecutive courses of chemotherapy for 45-60 patients (5-21 years of age). Patients with nausea/vomiting will be randomized to one of three groups: hypnosis, supportive counseling, and a no-treatment control. The groups will be matched at baseline for age and severity of nausea and vomiting. Patients in the first two more courses will be assessed without intervention. The control group will have all four courses assessed without intervention. Nurses who administer the chemotherapy will document anticipatory emesis in the clinic. Mean ratings of nausea and vomiting during each time period will be analyzed by a 3 (groups) x 3 (time periods) analysis of variance with repeated measures for time for each variable. Mean duration of nausea, vomiting, and disruption of eating, sleep, school, and activities during each of the 3 time periods will be analyzed by a Friedman analysis of variance (separately for the 3 groups) to determine if changes occurred over time. Comparisons across groups for each of the 3 times periods will be made by a Kruskal-Wallis analysis of variance. Parental and patients' ratings of severity of nausea and vomiting will be correlated (Pearson's r). If the frequency of anticipatory emesis changes for any of the groups, then the 3 groups would be compared at each course by a 2 x 3 chi-square to determine whether the 3 groups differed during baseline, intervention, and post-intervention courses. Hypnotic susceptibility scores will be divided (high vs. low scores) and compared (2 x 2 chi-square) to change scores for nausea and vomiting. The significance of this investigation is its potential for improving the quality of life for children with cancer.